© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 9541
    Prev
    Next
    INDEX
    Explanations

    formal legal terminology and phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    78
    +0.16
    1.0%
    52
    +0.10
    0.7%
    172
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    452
    +0.16
    0.01
    27
    +0.10
    0.01
    411
    +0.08
    0.01
    Negative Logits
    ctor
    -1.57
    merce
    -1.48
     etc
    -1.44
     center
    -1.42
    ques
    -1.39
    hots
    -1.34
    hold
    -1.32
    que
    -1.28
    atorium
    -1.27
    ments
    -1.26
    POSITIVE LOGITS
    Ī
    3.42
    °
    3.24
    Į
    3.24
    Ĩ
    3.17
    £
    3.16
    ĥ½
    3.11
    Ĵ
    3.10
    į
    3.09
    Ķ
    3.04
    ģ
    3.03
    Activations Density 0.006%

    No Known Activations